JP7183788B2 - Behavior determination device and behavior determination method - Google Patents

Description

The present invention relates to, for example, a technique for determining the behavior of a person who needs nursing care or care (hereinafter referred to as a person requiring nursing care, etc.).

In an aging society like our country (Japan) in recent years, it is expected that the number of people who need nursing care due to illness, injury, old age, etc. will further increase in the future.

Such persons requiring nursing care enter facilities such as hospitals and welfare facilities for the elderly, and receive nursing care. In such facilities such as hospitals and welfare facilities for the elderly, persons who provide nursing care to those requiring nursing care (hereinafter referred to as nurses, etc.) regularly make patrols to confirm their safety. However, compared to the day shift, the number of nurses and the like decreases during the semi-night shift and the night shift, and the work load per person increases. For this reason, in recent years, researches and developments have been made on monitored person monitoring devices that monitor a person requiring nursing care as a person to be monitored.

One of the techniques for monitoring a monitored person is a technique for determining the behavior of the monitored person. For example, in Patent Document 1, a bed area, a floor area, and a boundary side between these areas are set in an image, and when a human body is detected to move from the floor area to the bed area by passing through the boundary side, , a bed-entering event is determined, and a bed-leaving event is determined when a human body moves from a bed area to a floor surface area through a boundary side.

According to this technology, among the behaviors of the monitored person, it is possible to determine bed entry and bed exit.

The movement of the monitored person to leave the bed includes the movement of sitting on the side of the bed, putting his feet on the floor, and standing up. Some monitored persons take a long time to get out of bed. May appear to be in and out of bed repeatedly. When the monitored person makes such an action, according to the technique of Patent Document 1, there is a possibility that determinations of entering and leaving the bed are repeated (that is, an erroneous determination may occur).

JP-A-2002-230533

An object of the present invention is to provide a behavior determination device and a behavior determination method that can reduce erroneous determinations.

A behavior determination device according to a first aspect of the present invention includes an acquisition unit, an extraction unit, and a determination unit. The acquisition unit acquires a first image and a second image of a predetermined range including a monitored person to be monitored. The extraction unit extracts a human body region from the first image and the second image. The determining unit performs a first setting of setting, in the first image, a plurality of first regions of interest used for determining whether or not a predetermined first action has been performed, spaced apart by a distance. determining whether or not the monitored person has performed the first action based on the plurality of first attention areas and the human body area, and whether or not the monitored person has performed a predetermined second action; A second setting is performed in which a second region of interest used for determination of is set in the second image, and based on the second region of interest and the human body region, the monitored person determines the second region of interest. It is determined whether or not the action of

The above and other objects, features and advantages of the present invention will become apparent from the following detailed description and accompanying drawings.

1 is an explanatory diagram illustrating the configuration of a monitored person monitoring system according to an embodiment; FIG. FIG. 2 is a schematic diagram showing a living room in which a sensor device is arranged; It is a block diagram which shows the structure of a sensor apparatus. FIG. 10 is an explanatory diagram for explaining an image captured by the imaging unit when the person to be monitored is not in the room; FIG. 10 is an explanatory diagram for explaining an image in which an attention area is set; FIG. 4 is a plan view of an attention area used for determining whether or not a first action has been taken; FIG. 11 is a plan view of an attention area used for determining whether or not a second action has been taken; FIG. 10 is an explanatory diagram for explaining an image in which the monitored person is performing the first action; FIG. 11 is an explanatory diagram for explaining an image of a state in which the monitored person is performing the second action; FIG. 10 is an explanatory diagram illustrating a mode in which three attention areas used for determination of a first action are set; When the sensor device operates as an action determination device, it is a flow chart explaining this operation. It is explanatory drawing explaining the image imaged by the imaging part arrange|positioned diagonally above a bed. FIG. 10 is an explanatory diagram for explaining an image in which an attention area is set; FIG. 4 is an explanatory diagram for explaining frames (images) captured by an imaging unit when a person to be monitored is standing in front of a bed; FIG. 4 is an explanatory diagram for explaining frames (images) captured by an imaging unit when a monitored person sits on the side of a bed; FIG. 5 is an explanatory diagram for explaining a frame (image) captured by an imaging unit when a monitored person puts his/her legs in the bed; FIG. 10 is an explanatory diagram illustrating frames (images) captured by the imaging unit when the monitored person is performing an action that is assumed to have left the bed;

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments of the present invention will be described in detail based on the drawings. In each figure, the configurations denoted by the same reference numerals indicate the same configuration, and the description of the content of the configuration that has already been described will be omitted. In the present specification, reference numerals with suffixes omitted are used when referring to generically, and reference numerals with suffixes are used when referring to individual configurations.

A monitored person monitoring system according to an embodiment is a system for monitoring a monitored person who is a monitoring target to be monitored (in other words, a watching target person who is a watching target to be watched over) using a plurality of devices, and includes terminal devices. and a monitored person monitoring device that is communicably connected to the terminal device, detects a predetermined event related to the monitored person, and notifies the terminal device of the event. The monitored person monitoring device may be integrally configured as a single device. Each of the two types of devices is configured separately by providing a management server device that is configured as a separate device. The sensor device detects the predetermined event related to the monitored person and notifies (reports, transmits) the management server device. When the management server device receives the notification from the sensor device, the management server device manages the event for which the notification has been received and re-notifies (re-notifies, re-notifies) the event to a predetermined terminal device associated with the sensor device. Send. The terminal device may be one type of device, but in this specification, the terminal device is two types of devices, a fixed terminal device and a mobile terminal device. The main difference between these fixed terminal devices and mobile terminal devices is that the fixed terminal device is operated in a fixed manner, and the mobile terminal device is carried and operated by a supervisor (user) such as a nurse or a caregiver. is. Since the fixed terminal device and the mobile terminal device are substantially the same, the mobile terminal device will be mainly described below.

FIG. 1 is an explanatory diagram illustrating the configuration of a monitored person monitoring system MS according to an embodiment. More specifically, the monitored person monitoring system MS includes, for example, one or more sensor devices SU (SU-1 to SU-4), a management server device SV, a fixed terminal device SP, and one or more A mobile terminal device TA (TA-1, TA-2) and a private branch exchange (PBX, Private Branch Exchange) CX are provided. line) are communicably connected via NW. The network NW may be provided with repeaters such as repeaters, bridges and routers for relaying communication signals. In the example shown in FIG. 1, the plurality of sensor devices SU-1 to SU-4, the management server device SV, the fixed terminal device SP, the plurality of mobile terminal devices TA-1 and TA-2, and the private branch exchange CX are L2 switches. are communicably connected to each other by a wired and wireless mixed LAN (for example, a LAN conforming to the IEEE802.11 standard) NW including a line concentrator (hub, HUB) LS and an access point AP. More specifically, a plurality of sensor devices SU-1 to SU-4, a management server device SV, a fixed terminal device SP and a private branch exchange CX are connected to a line concentrator LS, and connected to a plurality of mobile terminal devices TA-1 and TA-2. is connected to the concentrator LS via an access point AP. The network NW constitutes a so-called intranet by using Internet protocols such as TCP (Transmission Control Protocol) and IP (Internet Protocol). The private branch exchange CX is connected to the telephone TL by the public telephone network PN.

The monitored person monitoring system MS is arranged at an appropriate place according to the monitored person Ob. The monitored person Ob is, for example, a person who needs nursing care due to illness or injury, a person who needs nursing care due to a decline in physical ability, a solitary person living alone, or the like. In particular, from the viewpoint of enabling early detection and early countermeasures, the monitored person Ob should be a person who needs to be found when a predetermined inconvenient event, such as an abnormal condition, occurs to him or her. preferable. Therefore, the monitored person monitoring system MS is preferably installed in buildings such as hospitals, welfare facilities for the elderly, and dwelling units according to the type of the monitored person Ob. In the example shown in FIG. 1, the monitored person monitoring system MS is installed in a building of a nursing care facility having a plurality of rooms RM where a plurality of monitored persons Ob reside and a plurality of rooms such as a nurse station.

The sensor device SU has a communication function or the like to communicate with other devices SV, SP, and TA via the network NW, detects a predetermined event related to the person to be monitored Ob, and reports the detected event to the management server device SV. , performs voice communication with the terminal devices SP and TA, generates images including moving images, and distributes the moving images to the terminal devices SP and TA. The predetermined events preferably include events requiring action.

As an example, FIG. 1 shows four first to fourth sensor units SU-1 to SU-4. The second sensor unit SU-2 is arranged in the room RM-1 (not shown) of Ob-1, and the second sensor unit SU-2 is arranged in the room RM-2 (not shown) of Mr. B Ob-2, who is one of the monitored persons Ob. The third sensor unit SU-3 is installed in the living room RM-3 (not shown) of Mr. C Ob-3 who is one of the monitored persons Ob, and the fourth sensor unit SU-4 is It is installed in the living room RM-4 (not shown) of Mr. D Ob-4 who is one of the monitored persons Ob.

The management server device SV has a communication function or the like to communicate with other devices SU, SP, and TA via the network NW. It is a device that manages information (monitoring information) regarding monitoring of Ob. When the management server device SV receives the first event notification communication signal as notification of the event from the sensor device SU, the management server device SV, based on each piece of information contained in the first event notification communication signal, determines the above information about the monitoring of the monitored person Ob. A predetermined terminal device that stores (records) monitoring information and transmits a communication signal (second event notification communication signal) containing the monitoring information relating to the monitoring of the monitored person Ob to the sensor device SU in advance. Send to SP, TA. For this reason, the management server device SV uses the sensor ID that is the transmission source and the notification A correspondence relationship (notification destination correspondence relationship) with a terminal ID that is a destination (re-notification destination) and its communication address are stored. A terminal ID (terminal device identifier) is an identifier for specifying and identifying the terminal devices SP and TA. The management server device SV then provides the clients (in this embodiment, the fixed terminal device SP, the portable terminal device TA, etc.) with data according to the request from the client. Such a management server device SV can be configured by, for example, a computer with a communication function.

The fixed terminal device SP is a device that functions as a user interface (UI) of the monitored person monitoring system MS. In order to achieve this function, the fixed terminal device SP has a communication function to communicate with other devices SU, SV, TA via the network NW, a display function to display predetermined information, and predetermined instructions and data. It has an input function, etc., for inputting predetermined instructions and data to be given to the management server device SV and the portable terminal device TA, and displaying monitoring information obtained by the sensor device SU. Such a fixed terminal device SP can be composed of, for example, a computer with a communication function.

The mobile terminal device TA is carried by the supervisor NS. The mobile terminal device TA has a communication function of communicating with other devices SV, SP, and SU via the network NW, a display function of displaying predetermined information, an input function of inputting predetermined instructions and data, and a voice call. It is equipped with a call function, etc., to input predetermined instructions and data given to the management server device SV and the sensor device SU, and to receive monitoring information (including moving images) obtained by the sensor device SU in response to a notification from the management server device SV. It is a device for displaying, responding to a nurse call, speaking to a nurse call, etc. by voice communication with the sensor unit SU.

In addition to the functions described above (for example, the function of capturing moving images of the monitored person Ob and the function of making voice calls with the terminal devices SP and TA), the sensor unit SU determines the behavior of the monitored person Ob. It has the function of a device (behavior determination device). From the point of view of this function, the sensor unit SU will be described in detail. FIG. 2 is a schematic diagram showing a living room RM in which the sensor device SU is arranged. A living room RM is provided with a bed 1 for a monitored person Ob. A bed 1 is a futon 5 spread on a bed 3. The bed 1 is not limited to this, and may be, for example, a futon 5 laid on tatami mats or a futon 5 laid on the floor. FIG. 2 shows a state in which the monitored person Ob is sleeping on the bed 1, and only the head of the monitored person Ob is shown in the body of the monitored person Ob. A sensor unit SU is attached to the ceiling 7 of the living room RM.

FIG. 3 is a block diagram showing the configuration of the sensor device SU (behavior determination device). The sensor device SU includes an imaging unit 10 , an extraction unit 11 , a determination unit 12 , a sensor-side control processing unit (SU control processing unit) 13 , and a sensor-side communication interface unit (SU communication IF unit) 14 . In FIG. 2, among the blocks constituting the sensor device SU, the imaging unit 10 is shown, and the other blocks are omitted.

2 and 3, imaging unit 10 is a device that captures an image of an object to be imaged and generates an image of the object to be imaged. Images include still images and moving images. The imaging unit 10 is arranged so as to be able to monitor the space in which the person to be monitored Ob is supposed to be located (located space, in the example shown in FIG. 1, the living room RM of the installation location). An image is captured to generate a bird's-eye view image of the imaging target. Preferably, since there is a high probability that the entire monitored person Ob can be imaged, the imaging unit 10 should be arranged so that the monitored person Ob's head is positioned on the bedding (for example, a bed) on which the monitored person Ob lies. It is arranged so that an image of the object to be imaged can be imaged from directly above a predetermined head position (usually, the position where a pillow is arranged). The sensor unit SU obtains an image of the monitored person Ob from above the monitored person Ob, preferably from right above the expected head position, using the imaging unit 10 .

Such imaging unit 10 may be a device that generates a visible light image, but in the embodiment, it is a device that generates an infrared image so that the monitored person Ob can be monitored even in relatively dark conditions. For example, in the embodiment, such an imaging unit 10 is an imaging optical system that forms an infrared optical image of an object to be imaged on a predetermined imaging plane, and is arranged with a light receiving surface aligned with the imaging plane. an image sensor that converts an infrared optical image of the object to be imaged into an electrical signal; and image data representing an infrared image of the object to be imaged by image processing the output of the image sensor. It is a digital infrared camera equipped with an image processing unit and the like for generating. In the embodiment, the imaging optical system of the imaging unit 10 is preferably a wide-angle optical system (a so-called wide-angle lens (including a fisheye lens)) capable of imaging the entire living room RM in which it is installed.

The imaging unit 10 is not limited to a camera such as an infrared camera, and may be, for example, a thermography or a three-dimensional sensor.

The imaging unit 10 functions as an acquisition unit that acquires a first image and a second image within a predetermined range including the monitored person Ob. The embodiment will be described using a moving image MI as an example of an image captured by the imaging unit 10, but the image is not limited to a moving image MI and may be a plurality of images arranged in time series.

The extraction unit 11 extracts a moving object region from each frame (image) forming the moving image MI captured by the imaging unit 10, and defines the extracted moving object region as a human body region. A moving object region is extracted from a frame by, for example, a technique such as background subtraction or optical flow.

The extraction unit 11 executes the functions of hardware such as a CPU (Central Processing Unit), a RAM (Random Access Memory), a ROM (Read Only Memory), and a HDD (Hard Disk Drive), and the extraction unit 11. It is realized by a program and data etc. for Some or all of the functions of the extraction unit 11 may be realized by processing by a DSP (Digital Signal Processor) in place of or together with the processing by the CPU. Also, similarly, some or all of the functions of the extraction unit 11 may be realized by processing by a dedicated hardware circuit in place of or together with processing by software. The above explanation also applies to the determination unit 12 and the SU control processing unit 13, which will be described later.

The determination unit 12 has a function of determining whether the monitored person Ob has taken the first action and a function of determining whether the monitored person Ob has taken the second action. Details of the determination unit 12 will be described later.

The SU communication IF unit 14 is a communication circuit that is connected to the SU control processing unit 13 and performs communication under the control of the SU control processing unit 13 . SU communication IF unit 14 generates a communication signal containing data to be transferred input from SU control processing unit 13 according to a communication protocol used in network NW of monitored person monitoring system MS, and transmits the generated communication signal. Signals are sent to other devices SV, SP, TA via the network NW. The SU communication IF unit 14 receives communication signals from other devices SV, SP, and TA via the network NW, extracts data from the received communication signals, and the SU control processing unit 13 can process the extracted data. format data and output to the SU control processing unit 13 . The SU communication IF unit 14 includes, for example, a communication interface circuit conforming to the IEEE802.11 standard or the like.

The SU control processing unit 13 controls each unit (the imaging unit 10, the extraction unit 11, the determination unit 12, and the SU communication IF unit 14) of the sensor unit SU according to the functions of the respective units, and controls a predetermined number related to the monitored person Ob. is detected, the detected event is notified to the management server device SV, voice communication is performed between the terminal devices SP and TA, and images including moving images are distributed to the terminal devices SP and TA. It is a device for

FIG. 4 is an explanatory diagram for explaining an image 30-1 captured by the imaging unit 10 when the person to be monitored Ob shown in FIG. 2 is not in the room RM. The imaging unit 10 is positioned directly above the bed 1 (FIG. 2). The imaging unit 10 captures the entire room RM. Image 30-1 includes an area representing bed 1 (below, bed area 22). The bedding area 22 has a rectangular shape.

A region of interest is set in the image 30-1. FIG. 5 is an explanatory diagram for explaining the image 30-1 in which the attention area 24 is set. The determination unit 12 stores position information of the attention areas 24 (attention areas 24-1 to 24-3) on the bed area 22 in advance. The setter (user of monitored person monitoring device MS, seller, etc.) views image 30-1 of living room RM including bed area 22 (bed image) and designates the position of bed area 22. FIG. Thereby, the determination unit 12 automatically sets the position of the attention area 24 on the image 30-1 and stores the position information. Using the position information, the determination unit 12 automatically sets a region of interest 24 in each frame (image) forming the moving image captured by the imaging unit 10 . Note that the configurator may set the position of the attention area 24 on the image 30-1 by viewing the image 30-1 and specifying the position of the attention area 24. 1, and correcting the position of the attention area 24 automatically set by the determination unit 12, setting may be made to correct the position of the attention area 24 on the image 30-1. The determination unit 12 stores the position information. Using the position information, the determination unit 12 automatically sets a region of interest 24 in each frame (image) forming the moving image captured by the imaging unit 10 .

Two attention areas 24 (attention areas 24-1 and 24-2) are set in the image 30-1. FIG. 6 is a plan view of the attention areas 24-1 and 24-2. Referring to FIGS. 5 and 6, attention areas 24-1 and 24-2 are examples of a plurality of first attention areas, and are used to determine whether or not the first action has been taken. As the first action, the monitored person Ob assumes a sleeping posture on the bed 1, but the first action is not limited to this. The monitored person Ob sleeping on the bed 1 may be in a sleep state or an awake state.

The attention areas 24-1 and 24-2 have a rectangular shape and are set along the longitudinal direction of the bed area 22 with a distance therebetween. The long sides of the attention areas 24 - 1 and 24 - 2 are parallel to the short sides of the bed area 22 . The short sides of the attention areas 24 - 1 and 24 - 2 are parallel to the long sides of the bed area 22 and are located on the long sides of the bed area 22 . The attention area 24-1 is an area including a portion where the monitored person Ob's head (in other words, pillow) is located. The region of interest 24-2 is a region including a portion of the futon 5 (FIG. 2) that covers the legs of the monitored person Ob.

A region of interest 24-3 is set in the image 30-1. FIG. 7 is a plan view of the attention area 24-3. 5 and 7, attention area 24-3 is an example of a second attention area, and is used to determine whether or not the second action has been taken. As the second action, an example in which the monitored person Ob puts a part of the body (legs) out of the bed 1 will be described, but the second action is not limited to this.

The attention area 24 - 3 has a rectangular shape and overlaps the bed area 22 . In FIG. 5, the line defining the attention area 24-3 is positioned on the line defining the bed area 22. In FIG. However, the invention is not limited to this, and the line defining the attention area 24-3 may be slightly deviated from the line defining the bed area 22. FIG. For example, the attention area 24-3 may be positioned slightly inside the bed area 22, or the attention area 24-3 may be positioned slightly outside the bed area 22. FIG.

An example in which a plurality of first attention areas (attention areas 24-1, 24-2) and a second attention area (attention area 24-3) are set in the same image 30-1 has been described. , may be set to different images.

3 and 5, determination unit 12 stores in advance position information indicating respective positions of attention areas 24-1, 24-2 and 24-3 on image 30-1. The position of the bed area 22 of each frame forming the moving image MI captured by the imaging unit 10 is the same as the position of the bed area 22 of the image 30-1. Therefore, when the determination unit 12 sets the attention areas 24-1, 24-2, and 24-3 for each frame, the settings are the same as those shown in FIG.

The determination unit 12 sets regions of interest 24-1, 24-2, and 24-3 for each frame of the moving image MI captured by the imaging unit 10, and extracts these regions of interest 24 and the extraction unit 11. Using the human body region thus obtained, it is determined whether or not the monitored person Ob has taken the first action (the monitored person Ob takes the posture of sleeping on the bed 1). It is determined whether or not the monitored person Ob has taken an action (a part of the body of the person to be monitored Ob has moved out of the bed 1). FIG. 8 is an explanatory diagram for explaining an image 30-2 in which the monitored person Ob is performing the first action. FIG. 9 is an explanatory diagram for explaining an image 30-3 in which the monitored person Ob is performing the second action. 8 and 9, the monitored person area 26 indicates the monitored person Ob. The monitored person area 26-1 shown in FIG. 8 corresponds to the head image of the monitored person Ob. The monitored person area 26-2 shown in FIG. 9 corresponds to an image in which the monitored person Ob sits on the side of the bed 3 (FIG. 2) with one leg out of the bed 3. FIG. The monitored person area 26-2 is composed of a monitored person area 26-21 located within the attention area 24-3 and a monitored person area 26-22 located outside the attention area 24-3.

A first action determination algorithm will be described. 3 and 8, the head of the monitored person Ob moves while the head of the monitored person Ob (that is, the monitored person area 26-1) is positioned in the attention area 24-1. Then, the extracting unit 11 extracts the region that has moved accordingly as the human body region. In this case, the monitored person area 26-1 is extracted as the human body area.

When the legs of the monitored person Ob are positioned in the region of interest 24-2 (the legs are under the futon 5 shown in FIG. 2) and the legs of the monitored person Ob move, the extraction unit 11 As a result, the moving area is extracted as the human body area. In this case, the portion of the futon 5 moved by the movement of the monitored person Ob's leg is extracted as the human body region, not the region indicating the leg.

Note that if the area of the human body region is small, there is a possibility of noise. Therefore, the determination unit 12 may determine the human body region using the following method. The determination unit 12 sets the human body region extracted by the extraction unit 11 as a human body region candidate. The determining unit 12 calculates the ratio of the area of the human body region candidate to the area of the attention region 24-1 (the number of pixels of the human body region candidate/the number of pixels of the attention region 24-1) for the human body region candidate located in the attention region 24-1. exceeds a predetermined first threshold. When this ratio exceeds the first threshold, the determination unit 12 determines the human body region candidate as the human body region. A similar process is performed for the human body area candidate located in the attention area 24-2. Note that the first threshold applied to the attention area 24-1 and the first threshold applied to the attention area 24-2 may be the same or different. If the optimum first threshold is the same for the attention area 24-1 and the attention area 24-2, it is the former, and if different, it is the latter.

When a person takes a sleeping posture on the bed 1, a human body area exists in each of the attention areas 24-1 and 24-2 (that is, in all of the attention areas 24-1 and 24-2). Therefore, the determination unit 12 determines that the monitored person Ob has taken the first action when the human body region exists in each of the attention regions 24-1 and 24-2.

In general, when a person enters the bed 1 and takes a sleeping position on the bed 1, the head and legs are not placed on the bed 1 at the same time, but the legs are first placed on the bed 1, and the head is placed on the bed 1 later. Put the part on bed 1. Therefore, after determining that the human body region (legs) exists in the attention region 24-2 corresponding to the legs, the determination unit 12 determines that the human body region (head) is present in the attention region 24-1 corresponding to the head. Determine that it exists. Considering the above, when the determination unit 12 determines that a human body region exists in each of the attention regions 24-1 and 24-2 within a predetermined period (for example, 10 seconds), the monitored person Ob is the first human body region. It is determined that the action of In other words, the determination unit 12 determines that one of the attention areas 24-1 and 24-2 has a human body area, and the attention areas 24-1 and 24 within a predetermined period (for example, 10 seconds) from this time. When it is determined that there is a human body area in the other of -2, it is determined that there is a human body area in each of the attention areas 24-1 and 24-2, and that the monitored person Ob has taken the first action.

An example has been described in which the attention area 24-1 is located in the area including the head of the monitored person Ob, and the attention area 24-2 is located in the area including the legs of the monitored person Ob. As long as it can be determined that Ob is in a sleeping posture, the position is not limited to these positions. For example, the attention area 24-1 may be located in an area including the head of the monitored person Ob, and the attention area 24-2 may be located in an area including the abdomen of the monitored person Ob. may be located in an area including the legs of the monitored person Ob, and the attention area 24-2 may be located in an area including the legs of the monitored person Ob.

Although the shapes of the attention areas 24-1 and 24-2 are rectangular, they are not limited to rectangles as long as it can be determined that the monitored person Ob is lying down. For example, the regions of interest 24-1 and 24-2 may have a square shape.

Although the attention areas 24-1 and 24-2 have the same area as an example, it is not limited to this as long as it can be determined that the monitored person Ob is in a sleeping posture.

Although two attention areas 24 (attention areas 24-1 and 24-2) are used for determining the first action, a plurality of attention areas may be used. The number is not limited to two, as long as it can be determined that the monitored person Ob is in a sleeping posture. A mode in which there are three attention areas 24 will be described as an example. FIG. 10 is an explanatory diagram illustrating a mode in which three attention areas 24 used for determination of the first action are set. Along the longitudinal direction of the bed area 22, an attention area 24-1, an attention area 24-4, and an attention area 24-2 are set at intervals. The attention area 24-4 is an area including the abdomen of the monitored person Ob. With the abdomen of the monitored person Ob positioned in the region of interest 24-4 (the abdomen is located under the futon 5 shown in FIG. 2), when the abdomen of the monitored person Ob moves, the extraction unit 11 (FIG. 3) extracts the moving region as the human body region.

The determination unit 12 determines that any one of the attention areas 24-1, 24-2, and 24-4 has a human body area, and determines that the remaining attention area is within a predetermined period (for example, 10 seconds) from this time. When it is determined that there is a human body area in the area 24, it is determined that the monitored person Ob has performed the first action.

A second action determination algorithm will be described. Referring to FIGS. 3 and 9, determination unit 12 determines whether or not there is a human body region within region of interest 24-3 and a human region continuous with this human body region is outside region of interest 24-3. judge. The monitored person area 26-21 is a human body area located within the attention area 24-3, and the monitored person area 26-22 is a human body area located outside the attention area 24-3.

For example, the determination unit 12 determines that the number of pixels of the monitored person area 26-22÷(the number of pixels of the monitored person area 26-21+the number of pixels of the monitored person area 26-22) exceeds the predetermined second threshold. When it exceeds, it is determined that the second action has been taken. The method of determining whether or not the second action has been taken is not limited to this. It may be determined that

The sensor device SU functions as an action determination device. The operation of the sensor device SU at this time will be described. FIG. 11 is a flow chart explaining this operation. 3 and 11, determination unit 12 sets "i" to "1" (step S1). “i” indicates the order of the frames forming the moving image MI captured by the imaging unit 10 . "1" means the first frame.

The determination unit 12 processes the first action determination algorithm for the first frame. Accordingly, the determination unit 12 determines whether or not the monitored person Ob has taken the first action when the first frame is captured (step S2). When the monitored person Ob enters the bed 1 (FIG. 2) and assumes the posture of lying on the bed 1, the determination unit 12 determines that the monitored person Ob has taken the first action.

When the determination unit 12 determines that the person to be monitored Ob is not taking the first action when the first frame is captured (No in step S2), the determination unit 12 replaces "i+1" with " i” (step S3). Here, "i" is set to "2".

The determination unit 12 processes the first action determination algorithm for the second frame. Accordingly, the determination unit 12 determines whether or not the monitored person Ob has taken the first action when the second frame is captured (step S2).

When the determination unit 12 determines that the monitored person Ob is not taking the first action at the time when the second frame is captured (No in step S2), the determination unit 12 replaces "i+1" with " i” (step S3). Here, "i" is set to "3". Thereafter, the determination unit 12 repeats the processing of steps S2 and S3 until the monitored person Ob takes the first action.

For example, when the determination unit 12 determines that the monitored person Ob has taken the first action when the j-th frame is captured (Yes in step S2), the determination unit 12 replaces "i+1" with " i” (step S4). Here, "i" is set to "j+1".

The determination unit 12 processes the second action determination algorithm for the i-th (j+1-th) frame. Accordingly, the determination unit 12 determines whether or not the monitored person Ob has taken the second action when the i-th (j+1-th) frame is captured (step S5).

When the determining unit 12 determines that the person to be monitored Ob is not taking the second action at the time when the i-th (j+1-th) frame is captured (No in step S5), the determining unit 12 i+1" is set to "i" (step S6). Here, "i" is set to "j+2". Thereafter, the determination unit 12 repeats the processing of steps S5 and S6 until the monitored person Ob takes the second action.

For example, when the determination unit 12 determines that the monitored person Ob has taken the second action when the k-th frame is captured (Yes in step S5), the determination unit 12 determines that the monitored person Ob It is determined that the user has woken up (step S7).

Main effects of the embodiment will be described. The embodiment focuses on this premise that when a person wakes up, he/she assumes a sleeping posture on the bed 1 (FIG. 2). 3 and 11, the determination unit 12 determines that the monitored person Ob is in a sleeping posture on the bed 1 (determines that the first action has taken place) (Yes in step S2), Next, when the monitored person Ob determines that a part of the body (for example, the leg) has moved out of the bed 1 (determined that the second action has taken place) (Yes in step S5), it is determined that the monitored person Ob has woken up. Determine (step S7).

When the monitored person Ob lying on the bed 1 wakes up, the monitored person Ob, as a series of actions, raises the upper body, sits on the side of the bed 1, and puts the feet on the floor. Then act to stand up. The action of sitting on the side of the bed 1 and putting the feet on the floor corresponds to the second action of the embodiment. Consider a method of determining that the person to be monitored Ob has woken up only by determining that the person to be monitored Ob has taken the second action. Among the persons to be monitored Ob, there are those who cannot smoothly sit on the side of the bed 1, put their feet on the floor, and stand up due to their weak muscles. Such a person shakes the body while sitting on the side of the bed 1, and uses this recoil to stand up. If the person to be monitored Ob behaves in this way, there is a possibility that the determination of waking up will be repeated only with the second behavior determination algorithm (that is, an erroneous determination may occur). Referring to FIG. 1, sensor device SU notifies mobile terminal device TA each time it determines that monitored person Ob has woken up. The device TA must be confirmed, which increases the burden on the supervisor NS.

On the other hand, in the embodiment, as described above, after it is determined that the monitored person Ob is sleeping on the bed 1 (determination that the monitored person Ob has taken the first action), the monitored person Ob is placed on the bed. From 1, it is determined that the monitored person Ob has woken up when it is determined that a part of the body has been taken out (determined that the second action has been performed). According to the embodiment, it is possible to reduce erroneous determinations of getting up as compared with the method of determining that the person to be monitored Ob has taken the second action, as compared with the method of determining that the monitored person Ob has taken the second action.

The embodiment has Modification 1 and Modification 2. FIG. Modification 1 will be described first. The embodiment determines that the monitored person Ob has woken up when the monitored person Ob performs the first action and then the second action. Modification 1 determines that the monitored person Ob has left the bed when the monitored person Ob performs the first action and then the second action.

In Modification 1, the imaging unit 10 shown in FIG. 2 is arranged diagonally above the bed 1 . FIG. 12 is an explanatory diagram for explaining an image 30-4 captured by the imaging section 10 arranged diagonally above the bed 1. As shown in FIG. Image 30-4 includes bedding area 22. FIG. The bed area 22 is shown stereoscopically. The image 30-1 shown in FIG. 4 shows the lines defining the image 30-1, but the image 30-4 shown in FIG. 12 omits the lines defining the image 30-4.

A region of interest is set in the image 30-4. FIG. 13 is an explanatory diagram for explaining the image 30-4 in which the attention area 24 is set. A region of interest 24-3 is set in the image 30-4. The attention area 24-3 is the same as the attention area 24-3 shown in FIG. The attention area 24-3 is used for determining whether or not the second action has been taken. In Modified Example 1, the second action is a motion that is estimated to have caused the monitored person Ob to get out of bed.

Three attention areas 24 (attention areas 24-5, 24-6, 24-7) are set in the image 30-4. The attention areas 24-5, 24-6, and 24-7 are examples of a plurality of first attention areas, and are used for determining whether or not the first action has been taken. In Modified Example 1, the first action is the action of the monitored person Ob entering the bed 1 .

The attention areas 24-5, 24-6, 24-7 have a rectangular shape, and the longitudinal direction of the attention areas 24-5, 24-6, 24-7 is the same as the longitudinal direction of the bed area 22. be. The attention areas 24-5, 24-6, 24-7 are set at intervals along the direction perpendicular to the longitudinal direction. A region of interest 24-6 is set between the region of interest 24-5 and the region of interest 24-7.

Region of interest 24-6 is an intermediate region of interest that overlaps bed region 22 in image 30-4. The attention area 24 - 6 overlaps the attention area 24 - 3 on the side of the bed area 22 . The region of interest 24-6 corresponds to the place where the monitored person Ob sits on the side of the bed 1. FIG. Region of interest 24-5 is the lower region of interest located below the middle region of interest in image 30-4. Region of interest 24-7 is the upper region of interest located above the middle region of interest in image 30-4.

FIG. 14 is an explanatory diagram for explaining a frame (image 30-5) captured by the imaging unit 10 when the monitored person Ob is standing in front of the bed 1. As shown in FIG. FIG. 15 is an explanatory diagram for explaining a frame (image 30-6) captured by the imaging unit 10 when the monitored person Ob sits on the side of the bed 1. As shown in FIG. FIG. 16 is an explanatory diagram for explaining a frame (image 30-7) captured by the imaging unit 10 when the monitored person Ob puts his/her legs in the bed 1. As shown in FIG. These images, like image 30-4 shown in FIG. 12, omit the lines defining the images. 14 to 16, the monitored person area 26-3 corresponds to the whole image of the monitored person Ob.

Referring to FIG. 14, an attention area 24-5 (lower attention area) and an attention area 24-6 (intermediate attention area) are displayed when the monitored person Ob stands in front of the bed 1. A region containing a portion of Ob's body. Here, the attention area 24-5 is an area including part of the leg of the monitored person Ob. The area where this area and the attention area 24-5 overlap is the monitored person area 26-31. The region of interest 24-6 is a region including part of the arm and part of the back of the monitored person Ob. The area where this area and the attention area 24-6 overlap is the monitored person area 26-32.

Referring to FIG. 15, an attention area 24-6 (intermediate attention area) and an attention area 24-7 (upper attention area) are displayed when the monitored person Ob sits on the side of the bed 1. It is a region that contains part of the body of the Here, the region of interest 24-6 is a region including part of the legs and waist of the monitored person Ob. The area where this area and the attention area 24-6 overlap is the monitored person area 26-33. The attention area 24-7 is an area including a portion of the monitored person Ob's head. The area where this area and the attention area 24-7 overlap is the monitored person area 26-34.

Referring to FIG. 16, an attention area 24-6 (intermediate attention area) is an area including a part of the monitored person Ob's body when the monitored person Ob puts his/her legs in the bed 1. As shown in FIG. Here, the region of interest 24-6 is a region including part of the arm and part of the leg of the monitored person Ob. The area where this area and the attention area 24-6 overlap is the monitored person area 26-35.

A first action determination algorithm will be described. With reference to FIGS. 3 and 14, when monitored person Ob moves closer to bed 1, extraction unit 11 extracts monitored person region 26-31 located in attention region 24-5 as a human body region. , the monitored person region 26-32 located in the attention region 24-6 is extracted as the human body region.

Referring to FIGS. 3 and 15, when monitored person Ob makes a motion to sit on the side of bed 1, extraction unit 11 converts monitored person region 26-33 located in region of interest 24-6 to human body region. , and the monitored person area 26-34 located in the attention area 24-7 is extracted as the human body area.

With reference to FIGS. 3 and 16, when monitored person Ob puts his/her legs into bed 1, extraction unit 11 converts monitored person region 26-35 located in region of interest 24-6 to human body region. Extract as

The method of extracting the human body region described with reference to FIGS. 14 to 16 is the same as in the embodiment.

The determination unit 12 first determines that the attention areas 24-5 and 24-6 have a human body area, and determines that the attention area 24-7 does not have a human body area (FIG. 14). It is determined that there is a human body area in the areas 24-6 and 24-7 and that there is no human body area in the attention area 24-5 (FIG. 15), and finally, there is a human body area in the attention area 24-6. Suppose that the determination is made and that there is no human body area in the attention areas 24-5 and 24-7 (FIG. 16). When the determination unit 12 makes these three determinations within a predetermined period (for example, 60 seconds), the determination unit 12 determines that the person to be monitored Ob made an action of entering the bed 1 (first action). do.

A second action determination algorithm will be described. FIG. 17 is an explanatory diagram for explaining a frame (image 30-8) captured by the imaging unit 10 when the person to be monitored Ob is performing an action presumed to have left the bed. The area where the monitored person Ob overlaps with the area outside the attention area 24-3 becomes the monitored person area 26-36 (the shaded area). For example, when the area (the number of pixels) of the monitored person regions 26 to 36 exceeds a predetermined fourth threshold value, the determination unit 12 determines that the monitored person Ob performs the second action (movement presumed to have left the bed). determine that you have done

Main effects of Modification 1 will be described. Modification 1 pays attention to the fact that when a person leaves the bed, the motion of entering the bed 1 is performed as a premise. In Modified Example 1, it is determined that the person to be monitored Ob has made an action (first action) to enter the bed 1, and then the person to be monitored Ob has taken an action (second action) that is presumed to have left the bed. When the determination is made, it is determined that the monitored person Ob has left the bed. According to Modification 1, it is possible to reduce erroneous determinations of getting out of bed, compared to the method of determining that the person to be monitored Ob has left the bed only by determining that the person to be monitored Ob has taken the second action.

Modification 2 will be described. Referring to FIG. 3, in the embodiment, an extraction unit 11 and a determination unit 12 are provided in the sensor device SU. On the other hand, in Modification 2, these are provided in the management server device SV shown in FIG. The SU control processing unit 13 provided in the modification commands the SU communication IF unit 14 to transmit the moving image MI captured by the imaging unit 10 to the management server device SV. The SU communication IF unit 14 converts the moving image MI into a communication signal that can communicate with a communication unit (not shown) of the management server device SV, and transmits the communication signal to the management server device SV.

In the embodiment, the imaging unit 10 functions as an acquisition unit that acquires a first image and a second image within a predetermined range including the monitored person Ob. On the other hand, in Modification 2, the communication unit of the management server device SV has the function of the acquisition unit.

(Summary of embodiments)
A behavior determination device according to a first aspect of an embodiment includes an acquisition unit that acquires a first image and a second image in a predetermined range including a monitored person to be monitored, and the first image and the an extraction unit that extracts a human body region from a second image; performing a first setting to be set on an image, determining whether or not the monitored person has performed the first action based on the plurality of first attention areas and the human body area; a second setting for setting a second region of interest used for determining whether or not a second action has been taken in the second image, and based on the second region of interest and the human body region; and a determination unit that determines whether or not the monitored person has taken the second action.

In the behavior determination device according to the first aspect of the embodiment, attention areas (a plurality of first attention areas set at intervals) used for determination of the first behavior and used for determination of the second behavior A region of interest (second region of interest) is separated. Therefore, it is possible to change the attention area to be set in the image according to the behavior to be determined. For example, the optimal attention area for determining the first action can be the attention area used for the first action determination, and the attention area optimal for the second action determination can be used for the second action determination. can be a region of interest used for Therefore, according to the behavior determination device according to the first aspect of the embodiment, erroneous determination can be reduced.

The first image and the second image may be the same image or different images.

In the above configuration, when the monitored person has determined that the monitored person has taken the second action after determining that the monitored person has taken the first action, the monitored person has taken a predetermined action. determined to have acted.

This configuration combines the first behavior and the second behavior to determine a predetermined behavior. This configuration has the following first and second aspects.

In the first aspect, the first image and the second image each include a bed area indicating a bed, and the first action is a posture in which the monitored person is sleeping on the bed. the second action is the monitored person taking out a part of the body from the bed; , when the human body region is set to be arranged in the bed region with a distance interval along the longitudinal direction of the bed region, and the human body region exists in each of the plurality of the first attention regions, the monitored person is in the first region of interest. 1 action is determined, and the determination unit sets the second attention area along the line defining the bed area as the second setting, and sets the second attention area outside the second attention area as the second setting. determining whether the monitored person has taken the second action based on the area of the human body region located at the and then, when it is determined that the monitored person has taken the second action, it is determined that the monitored person has woken up.

The first aspect focuses on the assumption that when a person wakes up, he/she assumes the posture of sleeping on the bed. The determination unit determines that the monitored person is in a sleeping posture on the bed (determines that the monitored person has taken the first action), and then the monitored person lifts a part of the body (for example, legs) from the bed. It is determined that the monitored person has woken up when it is determined that the monitored person has taken the second action. According to the first aspect, it is possible to reduce erroneous determination of waking up compared to a method of determining waking up only by determining that the monitored person has taken the second action.

When a person takes a sleeping posture on a bed, a human body area exists in each of the plurality of first attention areas (that is, in all of the plurality of first attention areas). Therefore, the determination unit determines that the monitored person has taken the first action when the human body region exists in each of the plurality of first attention regions. Normally, when a person enters the bed and assumes a sleeping position on the bed, the head and legs are not put on the bed at the same time, but one of them is moved in order or alternately little by little. Finally, both the head and legs are placed on the bed. Therefore, after determining that the human body region exists in at least one of the first attention regions, the determination unit determines that the human body region also exists in the remaining attention regions. Considering the above, preferably, when the determination unit determines that a human body region exists in each of the plurality of first attention regions within a predetermined period of time (for example, 10 seconds), the monitored person is in the first region of interest. determined to have acted.

There are various methods for determining whether or not the monitored person has taken the second action based on the area (the number of pixels) of the human body region located outside the second region of interest. is as follows. The determining unit divides the number of pixels in the human body region located outside the second region of interest/(the number of pixels in the human body region located in the second region of interest + the number of pixels in the human body region located outside the second region of interest). exceeds a predetermined threshold, it is determined that the second action has been taken.

In the second aspect, the first image and the second image include a bed area that indicates a bed, and the imaging unit that captures the first image and the second image captures the bed The first action is a motion of the monitored person getting into the bed, and the second action is an estimated motion of the monitored person leaving the bed. , the plurality of first regions of interest include a middle region of interest overlapping the bed region in the first image, a lower region of interest positioned below the middle region of interest in the image, and the middle region of interest in the image. The determination unit sets the lower attention area, the intermediate attention area, and the upper attention area as the first settings, and the lower attention area is configured by an upper attention area positioned above the lower attention area. a first state in which the human body region exists in the region of interest and the intermediate region of interest and the human body region does not exist in the upper region of interest; a second state in which the human body region does not exist in the lower region of interest; When it is determined that the nonexistent third state occurs in the order of the first state, the second state, and the third state, it is determined that the monitored person has taken the first action. and, as the second setting, the determination unit sets the second attention area along a line defining the bed area, and determines the human body area located outside the second attention area. Based on the area, it is determined whether or not the monitored person has taken the second action, and after the determination unit has determined that the monitored person has taken the first action, the monitored person When it is determined that the monitored person has taken the second action, it is determined that the monitored person has left the bed.

The second mode focuses on the motion of getting into bed as a premise when a person leaves the bed. In the second mode, it is determined that the monitored person has made a motion of going into bed (first behavior), and then it is determined that the monitored person has made a behavior that is presumed to have left the bed (second behavior). Then, it is determined that the monitored person has left the bed. According to the second aspect, it is possible to reduce erroneous determinations of getting out of bed compared to a method of determining that the monitored person has left the bed only by determining that the person to be monitored has taken the second action.

There are various methods for determining whether or not the monitored person has taken the second action based on the area (the number of pixels) of the human body region located outside the second region of interest. is as follows. The determination unit determines that the second action is taken when the number of pixels of the human body region located outside the second region of interest exceeds a predetermined threshold.

A behavior determination method according to a second aspect of an embodiment includes an acquisition step of acquiring a first image and a second image in a predetermined range including a monitored person to be monitored; an extraction step of extracting a human body region from a second image; performing a first setting to be set on an image, determining whether or not the monitored person has performed the first action based on the plurality of first attention areas and the human body area; a second setting for setting a second region of interest used for determining whether or not a second action has been taken in the second image, and based on the second region of interest and the human body region; and a determination step of determining whether or not the monitored person has taken the second action.

The behavior determination method according to the second aspect of the embodiment defines the behavior determination device according to the first aspect of the embodiment from the viewpoint of the method, and is similar to the behavior determination device according to the first aspect of the embodiment. has the action and effect of

This application is based on Japanese Patent Application No. 2016-137422 filed on July 12, 2016, the content of which is included in the present application.

Although the present invention has been adequately and fully described above through embodiments with reference to the drawings in order to express the present invention, modifications and/or improvements to the above-described embodiments can easily be made by those skilled in the art. It should be recognized that it is possible. Therefore, to the extent that modifications or improvements made by those skilled in the art do not depart from the scope of the claims set forth in the claims, such modifications or improvements do not fall within the scope of the claims. is interpreted to be subsumed by

According to the present invention, it is possible to provide a behavior determination device and a behavior determination method.

Claims (3)

an acquisition unit arranged on a ceiling for acquiring a first image and a second image of a predetermined range including a monitored person to be monitored;
an extraction unit that extracts a human body region from the first image and the second image;
performing a first setting of setting a plurality of first regions of interest used for determining whether or not a predetermined first action is performed in the first image at intervals of a distance; Based on the first region of interest and the human body region, it is determined whether or not the monitored person has taken the first action, and is used to determine whether or not the person has taken a predetermined second action. A second setting is performed to set a second attention area to be observed in the second image, and the monitored person performs the second action based on the second attention area and the human body area. A determination unit that determines whether or not,
The determination unit determines that the monitored person has taken a predetermined action when it is determined that the monitored person has taken the second action after determining that the monitored person has taken the first action. judge ,
The first image and the second image include a bed area, which is an area indicating a bed ,
the first action is that the monitored person assumes a sleeping position on the bed ;
the second action is the monitored person leaving a part of the body from the bed ;
The determination unit sets, as the first setting, the plurality of first attention areas to be arranged in the bedding area with distance intervals along the longitudinal direction of the bedding area, and determining that the monitored person has taken the first action when the human body region exists in each of the attention regions of
As the second setting, the determination unit sets the second region of interest along a line defining the bed region, and determines the area of the human body region located outside the second region of interest. determining whether the monitored person has taken the second action based on
The determination unit determines that the monitored person has woken up when it is determined that the monitored person has performed the second behavior after determining that the monitored person has performed the first behavior. Action judgment device. an acquisition unit that acquires a first image and a second image of a predetermined range including a monitored person to be monitored;
an extraction unit that extracts a human body region from the first image and the second image;
performing a first setting of setting a plurality of first regions of interest used for determining whether or not a predetermined first action is performed in the first image at intervals of a distance; Based on the first region of interest and the human body region, it is determined whether or not the monitored person has taken the first action, and is used to determine whether or not the person has taken a predetermined second action. A second setting is performed to set a second attention area to be observed in the second image, and the monitored person performs the second action based on the second attention area and the human body area. A determination unit that determines whether or not,
The determination unit determines that the monitored person has taken a predetermined action when it is determined that the monitored person has taken the second action after determining that the monitored person has taken the first action. judge,
The first image and the second image include a bed area, which is an area indicating a bed,
an imaging unit that captures the first image and the second image is arranged diagonally above the bed,
the first action is an action of the monitored person entering the bed;
the second action is a motion presumed to have caused the monitored person to get out of bed;
The plurality of first attention areas includes an intermediate attention area overlapping the bed area in the first image, a lower attention area positioned below the intermediate attention area in the image, and an intermediate attention area in the image. It is composed of an upper region of interest located above,
The determination unit sets the lower attention area, the intermediate attention area, and the upper attention area as the first settings, and the human body area exists in the lower attention area and the intermediate attention area. and a first state in which the human body region does not exist in the upper region of interest, the human body region exists in the intermediate region of interest and the upper region of interest, and the human body region exists in the lower region of interest. and a third state in which the human body region exists in the intermediate region of interest and no human body region exists in the lower region of interest and the upper region of interest are the first state. , when determining that the second state and the third state have occurred in this order, determining that the monitored person has performed the first action;
As the second setting, the determination unit sets the second region of interest along a line defining the bed region, and determines the area of the human body region located outside the second region of interest. determining whether the monitored person has taken the second action based on
The determination unit determines that the monitored person has left the bed when determining that the monitored person has performed the second behavior after determining that the monitored person has performed the first behavior. judgment device. an acquisition step of acquiring a first image and a second image of a predetermined range including a monitored person to be monitored by a camera arranged on the ceiling;
an extracting step of extracting a human body region from the first image and the second image;
performing a first setting of setting a plurality of first regions of interest used for determining whether or not a predetermined first action is performed in the first image at intervals of a distance; Based on the first region of interest and the human body region, it is determined whether or not the monitored person has taken the first action, and is used to determine whether or not the person has taken a predetermined second action. A second setting is performed to set a second attention area to be observed in the second image, and the monitored person performs the second action based on the second attention area and the human body area. and a determination step of determining whether
The first image and the second image include a bed area, which is an area indicating a bed ,
the first action is that the monitored person assumes a sleeping position on the bed ;
the second action is the monitored person leaving a part of the body from the bed ;
In the determining step, as the first setting, a plurality of the first attention areas are set to be arranged in the bedding area at intervals along the longitudinal direction of the bedding area; determining that the monitored person has taken the first action when the human body region exists in each of the attention regions of
In the determination step, as the second setting, the second region of interest is set along a line defining the bed region, and the area of the human body region located outside the second region of interest is determined. determining whether the monitored person has taken the second action based on
The determining step determines that the monitored person has woken up when it is determined that the monitored person has performed the second behavior after determining that the monitored person has performed the first behavior. Action judgment method.

Images